Browsing by Author "Barrios Herrera, Lizandra"

Now showing 1 - 1 of 1
  • Thumbnail Image
    ItemOpen Access
    Investigation of Ni-Ceria Nanocatalysts: Improved Computational Methods Towards Hydrogen Production in Aquaprocessing of Heavy Oil
    (2023-02-21) Barrios Herrera, Lizandra; Salahub, Dennis; Kusalik, Peter; Pereira Almao, Pedro
    Ni-Ceria (Ni-CeO2) nanoparticles (NPs) are promising nanocatalysts for water splitting and water gas shift reactions due to the ability of ceria to temporarily donate oxygen to the catalytic reaction and accept oxygen after the reaction is completed. Therefore, elucidating how different properties of the nanoparticles, such as structure, size, composition, and electronic structure, relate to the activity and selectivity of the catalytic reaction is crucial for developing novel catalysts. In this thesis, we focus on the computational investigation of Ni-Ceria NPs to shed light on their structure/property relationships. From the fundamental point of view, the computational investigation of Ni-CeO2 systems presents a high degree of complexity because they are characterized by strong electron correlation, and it is known that first-principles calculations, such as Density Functional Theory (DFT), may fail to treat the correlation properly. To fill that gap, and with the motivation of increasing the functionality of the deMon2k software, we implemented the Hubbard model, commonly known as +U, in the framework of Auxiliary Density Functional Theory (ADFT) to treat the correlation explicitly. The formulation and implementation of our “ADFT+U” model are discussed in Chapter 3. On the other hand, in Chapter 4, we introduce new auxiliary functions for pseudo-potentials addressing the shortcomings of the auxiliary functions in previous versions of deMon2k, and validate ADFT for Ni-Ceria systems. It is known that the global optimization of nanoparticles characterized by complex potential energy surfaces with many local minima, such as Ni-Ceria NPs, remains challenging and time-consuming. In Chapter 5, we use machine learning (ML) regression and its uncertainty -- an active-learning (AL) strategy -- for the global optimization of the nanoparticles using DFT calculations. The method allows the learned structure-energy relationships to improve iteratively when more data are obtained from DFT calculations of promising structures indicated by the AL. Additionally, further investigation of the NPs by mass-scaled parallel-tempering Born-Oppenheimer molecular dynamics (PT-BOMD) resulted in the same global minima (GM) structures found by AL, demonstrating the robustness of our AL search. This AL strategy has shown to be a powerful tool for learning from small datasets and assists in the global optimization of complex electronic structure systems, such as Ni-Ceria NPs, allowing us to determine the structures of Ce(y-x)NixO(2y-x), with (x=1, 2, 3; y=4) and Ce10Ni3O26 NPs for the first time. Finally, in Chapter 6, we present a novel AL model to search for the catalytic sites of the NPs based on single-point calculations, which is an essential step to explore the reaction over the nanoparticle surface. In summary, this thesis provides crucial insights into the relationships between the structures, the electronic structures, and the catalytic properties of Ni-Ceria NPs for the water-splitting reaction. For instance, we found that Ni-Ceria NPs with a 3Ni/1Ce ratio exhibit a more favourable adsorption energy, which relates to the electronic properties of the cluster, with a computed HOMO-LUMO gap of about 2 eV, and so, better catalytic properties for the water-splitting reaction.